rejected. This occurred very infrequently and was due mainly to failure to 

 allow the rate of feed of compressed air to become constant before beginning 

 the test. The change in weight due to the bubble gives the volume of the 

 water displaced, therefore the volume of the bubble or a counted number of bub- 

 bles is known. The volume so obtained must be adjusted because the pressure 

 at which the bubble size is measured is different from that at which the rate 

 of rise is determined. The effect of surface tension on the pressure inside 

 the bubble is negligible. 



V i " (1^-p9 V o 18] 



where V is the volume of the bubble when it is in the field of the camera, 



V is the volume of the bubble as determined by weighing, 



P is the absolute pressure at the funnel, 



P is the pressure at the camera level, and 



P 1 is the vapor pressure of water at the test temperature. 



Since the water in the tank had been exposed to the atmosphere for 

 a considerable length of time, it could be assumed saturated with air. Conse- 

 quently, no consideration need be given to the possibility of decrease in the 

 size of the bubble due to its dissolving in the water as it rose. 



Most of the bubbles below an r of 0.06 cm were released through 

 fine hypodermic needles and could not be generated with a uniform size. The 

 sizes of these bubbles were determined with the comparator. This method was 

 compared with the gravimetric method for several bubbles and found to give 

 agreement within a few percent. Since the bubbles used for comparison pur- 

 poses were somewhat larger than those released from the hypodermic needles, 

 larger errors in determining the size of the smallest bubbles could be 

 expected. 



The rate of rise of the bubbles was determined by measuring the dis- 

 placement of the bubbles in successive frames of the film. Figure 3 shows the 

 displacement as a function of the time for several typical runs. The result- 

 ant curves are straight lines indicating that the velocity of the bubble is 

 constant during the interval it traverses the camera field. 



DISCUSSION OF RESULTS 



Since bubbles of different size will assume different shapes, the 

 equivalent radius r , defined as the radius of a sphere having the same volume 

 as that of the bubble, was used as the length parameter. Figure 4 shows the 

 terminal velocity of air bubbles in water at 67 F as a function of the 



